Hyperspectral Image Classification: Potentials, Challenges, and Future Directions

Abstract

Recent imaging science and technology discoveries have considered hyperspectral imagery and remote sensing. The current intelligent technologies, such as support vector machines, sparse representations, active learning, extreme learning machines, transfer learning, and deep learning, are typically based on the learning of the machines. These techniques enrich the processing of such three-dimensional, multiple bands, and high-resolution images with their precision and fidelity. This article presents an extensive survey depicting machine-dependent technologies’ contributions and deep learning on landcover classification based on hyperspectral images. The objective of this study is three-fold. First, after reading a large pool of Web of Science (WoS), Scopus, SCI, and SCIE-indexed and SCIE-related articles, we provide a novel approach for review work that is entirely systematic and aids in the inspiration of finding research gaps and developing embedded questions. Second, we emphasize contemporary advances in machine learning (ML) methods for identifying hyperspectral images, with a brief, organized overview and a thorough assessment of the literature involved. Finally, we draw the conclusions to assist researchers in expanding their understanding of the relationship between machine learning and hyperspectral images for future research.

Hyperspectral imaging for early diagnosis of diseases: A review

“…Accuracy is a crucial aspect when dealing with medical diagnosis. Classical state of the art machine learning techniques suffer from the problem of less accurate band extraction and feature selection in HSI (Datta et al, 2022). Noise is also a serious hindrance to SOTA methodologies.…”

Curvelet Transform Based Compression Algorithm for Low Resource Hyperspectral Image Sensors

“…Te pseudo code for the 3D-LEZSPC is given as (See Algorithm 1). Te coefcient is not examined signifcantly for the current bit plane α [1] Te coefcient will be examined for signifcance in the current bit plane α [2] Te coefcient is newly signifcant to the current bit plane α [3] Te coefcient was signifcant in the previous bit plane Te tree node is not examined signifcantly for the current bit plane…”

“…Due to the recent advancement in imaging spectrometry, the HS image sensors capture the refectance (intensity) of the scene with increasingly higher spatial and spectral resolution [1]. Tis abundant information makes HSI useful in multiple applications from remote sensing, anomaly detection, land surface classifcation, and environmental monitoring [2].…”

Noisy and Mixed Pixel Aware Hyperspectral Crop Classification

“…Hyperspectral remote sensing is very much an essential approach for obtaining multiple incredibly small spectrally continuous imaging information in the electromagnetic spectrum's visible, thermal infrared, near-infrared, mid-infrared bands. The comprehensive spectral reflection properties allow for reliable differentiation (classification) of the vegetation cover of concern [1], [2]. Several HSI classification algorithms are developed in order to suit the various needs in vegetative monitoring and assessment, agricultural determination, diseases and pests' detection, and practical surveillance [3], [4].…”